Lloyd uninitentionally brought up a good point.
MPI_INTERCOMM_PARTITION was dropped in September, I will I was unavailable
to defened it. This was the important call to allow intracomms to be
split into intercommunicators. Without it, we have only the more clumbsy,
less synchronous MPI-1 call, which is also more complex to use, and
unnecessarily much work for a user, who just wants to divide a SPMD world.
Applications will have to work unnecessarily hard with current spec.
Since a fully formed proposal was offered previously, and was discarded
most probably for lack of motivation, I suggest it be reconsidered
at the next meeting. The text was available before, and will be
rerouted to Bill Gropp.
This is not a performance-oriented call, but a convenience-oriented call.
It is very similar in vein to MPI_COMM_SPLIT, and was on the collective
chapter books for more than 1.5 years. It does not adversely impact
implementations.
Tony
On Sat, 25 Jan 1997, Tony Skjellum wrote:
> Date: Sat, 25 Jan 1997 18:46:05 -0600 (CST)
> From: Tony Skjellum <tony@Aurora.CS.MsState.Edu>
> To: "Lewins, Lloyd J" <llewins@msmail4.hac.com>
> Cc: mpi-coll <mpi-coll@mcs.anl.gov>
> Subject: Re: 6.4 Intercommunicator Constructors
>
> That is a remaining orphan text. That constructor was deleted. The
> text should go. -Tony
> On 24 Jan 1997, Lewins, Lloyd J wrote:
>
> > Date: 24 Jan 1997 08:30:43 -0800
> > From: "Lewins, Lloyd J" <llewins@msmail4.hac.com>
> > To: mpi-coll <mpi-coll@mcs.anl.gov>
> > Subject: 6.4 Intercommunicator Constructors
> >
> > Page 118, line 11 states that "One other intercommunicator constructor for
> > partitioning intracommunicators into multiple intercommunicators is proposed".
> > I do not see this conctructor!!
> >
> > Lloyd Lewins
> > Hughes Aircraft Co.,
> > llewins@msmail4.hac.com
> >
>
> Anthony Skjellum, PhD, Associate Professor of Computer Science;
> Mississippi State University, Department of Computer Science & NSF ERC
> Butler, Rm 300, PO Box 9637, Corner of Perry&Barr, Mississippi State,MS 39762
> (601)325-8435 FAX: (601)325-8997; http://www.erc.msstate.edu/~tony;
> "Persistence is fertile." ; e-mail: tony@cs.msstate.edu; Try MPI!
>
>
Anthony Skjellum, PhD, Associate Professor of Computer Science;
Mississippi State University, Department of Computer Science & NSF ERC
Butler, Rm 300, PO Box 9637, Corner of Perry&Barr, Mississippi State,MS 39762
(601)325-8435 FAX: (601)325-8997; http://www.erc.msstate.edu/~tony;
"Persistence is fertile." ; e-mail: tony@cs.msstate.edu; Try MPI!